69

BREEDING METHODS AND PROCEDURES EMPLOYED AT IRRI DEVELOPING RICE GERM PLASM WITH

RESISTANCE TO AND INSECTS

Gurdev S. KHUSH *

Rice is the host of over 60 diseases and more than 1 00 insect pests. Some of these are of major international importance. During the last ten years major changes have occurred in the varietal composition and cultural practices for rice. High yielding varieties are now planted in approximately 50% of the 130 million hectares occupied by rice all over the world. These varieties are characterized by early maturity, photoperiod insensitivity, short stature, high tillering, and dark green erect leaves. A relatively small number of improved varieties have literally hundreds of traditional thereby reducing the genetic variability of the crop.

In the wake of the introduction of varieties with improved plant type, farmers have started using improved cultural practices such as the application of more fertilizers and the establishment of higher plant populations per unit area. Development of irrigation facilities and availability of early maturing, photoperiod-insensitive varieties have enabled the farmers in tropical Asia to grow successive crops of rice throughout the year covering large areas.

Reduced genetic variability, improved cultural practices, and continuous cropping with intended for increased rice production, have increased the genetic vulnerability of the

crop. Within the last few years serious outbreaks of diseases and insect pests have affected ride in several countries. Very little research has been done on the chemical control of rice diseases in the tropics. Chemical control of high insect populations for prolonged periods in tropical climates where insect generations overlap throughout the year is very expensive. Social and economic conditions in the tropics present other obstacles to the chemical control of rice diseases and insects. The use of host resistance to control diseases and insects is the most logical approach to overcome these production constraints. Therefore, major emphasis has been placed at IRRI on developing germ plasm with multiple resistance to major diseases and insects. Varieties and breeding lines with resistance to as many as five diseases and five insect species have been developed. This paper describes the breeding methods and procedures employed in developing disease and insect resistant rice.

Major diseases and insects of rice In most of the rice growing areas, more than one disease or insect cause serious yield

losses. In Latin America, for example, blast, hoja blanca, and Sogatodes oryzicola are the factors limiting rice production. In Africa, blast and stemborers take serious toll of rice yields. In Asia, where 92% of the rice is produced and consumed, more than a dozen diseases and insects cause losses of epidemic proportions. In tropical Asia, disease and insect problems are most serious because of year round favorable climate and long history of rice cultivation conducive to the development of many diseases and pest organisms. One year there may be an epidemic of bacterial blight, the next year green leafhopper and tungro may cause serious damage and the following year an outbreak of brown planthopper and grassy stunt may be observed. Therefore, to minimize the yield losses from diseases and insects, varieties with multiple resistance to most of the major diseases and insects are required. Five diseases (blast,

* Plant Breeder, International Rice Research Institute, P.O. Box 933, Manila, The Philippines.

70

The

Fig. l.

program was

10081 e''~ Rosislml 1o 4

Bf. Resiston! to 5

-fteoio- "'6

!969 1970 197! 1972 1973 1974 1975

made amongst

Changes in proportion of entries in IRRI's annual replicated yield trials with multiple resistance to important diseases and insects in the Philippines (blast, bacterial blight, tungro, grassy stunt, brown planthopper, and green leafhopper). Each year's trial consisted of at least 185 entri,,s,

the national programs. This used for commercial

the resistant lines as varieties under the names of

of IRRl named varieties and IRRI lines in Table . The

to IR42 is evident. increase in

and IR42. The disease and insect named the Government level cif resistance of varieties from IR8

!R5 !R8 !R20 !R22 !R24 IR26 !R28 !R29 IR30 lR32 lR34 !R36* !R38** IR40** IR42*

Table Disease and in.~ect resistance reactions varieties named by IRRI and those named by the Philippine Government.

'\1R s

MR s s

MR R

R MS MR

R R R R R

Bacterial Blight

s s R R s R R R R

R R R R R

Grassy Stunt

s

s

s MS

R R R R R R R R R

Disease and Insect Reactions*

Tungro

s MR s s

MR R R

MR MR

R R R R R

Leaf· hopper

R R R

R R R R R

R R R R R R

Brown Plant· hopper

s s s s ,. ,)

R R R R R R R R R R

Sh~ln­

borer

MS

MR s s

MR MR MR MR MR MR MR MR MR MR

S- Susceptible; MS Moderately susceptible; MR ·Moderately resistant; R ·Resistant. Reactions based on tests eondULoted in the Philippines for all diseases und insects except gall midge. Screening for gall midge was done in India.

** - Named bv the Philippine Government.

Gall Midge

s

s s s s s s s R

s R R R R

72

Breeding methods employed We have almost exclusively the method of for

the multiple resistant germ plasm. With this method the selection can be based on records of the disease and insect reactions of each line and, in the case of F4 and later generation

on the reaction of ancestral lines as welL The bulk method is not used as it does not permit the concurrent screening for a number of diseases and insects. The backcross method has not been employed extensively because of a lack of suitable recurrent parents. A few backcrosses were made in the crosses with 0. nivara for grassy stunt resistance; IRS and IR24 were used as recurrent parents. After three or four backcrosses, we obtaine·d grassy stunt resistant breeding lines similar to the recurrent parents, but they lacked resistance to other important diseases and insects such as tungro, bacterial blight, and brown planthopper. These breeding lines were again used in hybridization programs as sources of grassy stunt resistance. However, numerous suitable recurrent parents with multiple resistance, including the named varieties such as IR28, IR32, IR34, IR36, IR38, IR40, and IR42 are now available and are being used as recurrent parents in a backcrossing program to transfer resistance to white backed planthopper from N22 and Dharial.

We are also using backcross method to develop isogenic lines with different genes for resistance to brown planthopper.

Pedigree method of breeding is eminently suited to disease and insect resistance programs if the resistance is governed by major genes. In rice, resistance to blast (Kiyosawa, 1974), bacterial blight (Khush, 1977), tungro (Shastry et al., 1972), grassy stunt (Khush and Ling, 1974), green leafhopper (Athwal et 1971; Siwi and Khush, 1977), brown planthopper (Athwal et a!., 1971; Lakshminarayana and Khush, 1977), and gall midge (Satyanarayanaiah and Reddi. 1972) is under major gene control. Therefore, it was possible to combine genes for resistance to six or seven major diseases and insects together in such a short period.

For traits governed by polygenes, the pedigree method of breeding is not so suitable. Resistance to stemborers and sheath blight appears to be under polygenic control. For these two traits we are using a diallele selective mating system proposed by Jensen (1970). This method involves: (1) crossing a number of moderately resistant parents in all possible com­binations: (2) intercrossing the F 1 populations so obtained in all the possible combinations; (3) screening the double-cross F 1 progeny for resistance: and ( 4) intercrossing the selected plants found to have better resistance than either of the parents. The crossing, screening, selection and recrossing will be continued until minor genes from different sources are accumulated and the intensity of the trait is built up.

We are also exploring the feasibility of employing the rapid generation advance method for improving the traits governed by polygenic variation. Early generation populations from multiple crosses involving three or four parents with minor gene resistance are crossed and propagated in bulk using the rapid generation advance method. With this method it is possible to grow 3 - 4 generations in a year. No selection is practiced during this period. At F 5 or F6

the bulk population is exposed to the disease or insect pressure and individuals with better level of resistance are identified and grown in progeny rows for further evaluation.

Breeding procedures As a result of ten years of experience on breeding for disease and insect resistance we

have developed a set of procedures for handling the donor parents, making the crosses, growing and screening of segregating populations and concurrent evaluation of the materials for agronomic traits and grain quality. These procedures are brietly reviewed here.

Donor parents. Each season we plant the donor parents into a hybridization block. The hybridization block consists of 200 - 250 entries and includes newly identified unimproved donor parents as well as breeding lines with resistance to specific diseases and insects. The

entries of the hybridization intervals. This schedule insures times during the crossing season.

73

each season at bi-weekly crossing work at different

Hybridization. A number of crosses are made each season. in producing the single cross F 1 hybrids, each donor parent or breeding line is crossed with a number of other breeding lines. Thus, a set of single cross F 1 progenies is available for making the double or top-crosses in the next season. All the F 1 's involving the same donor parent or breeding line are grown together in the F 1 nursery. The best ones are used for making the top-crosses or double crosses. The top-cross parent is selected to complement the deficiency of the single cross F 1 hybrid. Thus, if one parent of the single cross F 1 is resistant to blast and bacterial blight and the other is resistant to tungro, the top-cross parent should have resistance to brown planthopper and green lealbopper. It is desirable to use the top-cross parents which are homozygous for resistance. All the F 1 plants then inherit the trait. It is also desirable to use improved plant type breeding line or variety as top-cross parent. If the unimproved tall donor is used as a top-cross parent all the F 1 progenies of the top-cross are tall and only Yi of the F 2 progeny would be of short stature.

In making a top-cross or double cross a fairly large number of F 1 seeds (300 · 400) are obtained. This allows us to sample gametic variability of the single cross F 1 hybrids. The screening starts from the F 1 generation. Let us consider a double cross between four parents of which A is resistant to bacterial blight, B is resistant to grassy stunt, C is resistant to brown planthopper, and D is resistant to green leafhopper. All these traits are under the control of dominant genes that segregate independer,tly of each other. About 400 seeds from the double cross A/B/ / C/D are obtained. The seeds are germinated and inoculated with grassy stunt virus in the greenhouse. Approximately 50% of the seedlings are susceptible and are elimi­nated. The remaining 200 are transplanted in the field and inoculated with bacterial blight at the age of about 60 days. About 50% of these which are susceptible are destroyed. The remaining 100 plants are harvested separately. Two small seed s.amples are taken from each and are progeny-tested for resistance to brown planthopper and green leafhopper. Those carrying the brown planthopper resistance gene (50%) and the ones carrying the green leafhopper resistance gene (50%) are identified. The F 2 populations are grown only from those carrying both genes (25 to 30 plants). Thus, by judicious, and timely screening, the original F 1 sample of 400 is reduced to 25 - 30 plants. All the F 2 populations grown from these plants segregate for the four resistance genes.

F 2 populations. Most of the F 2 populations are grown without insecticide protection and are thus exposed to the onslaught of leafhoppers and plant hoppers. Generally, enough inoculum of the virus diseases is present on the IRRI farm and we get high disease incidence in the F 2 populations. It is not uncommon to see all the F 2 plants of certain cross combinations infected with tungro virus. These populations are discarded .. Most of the F 2 populations are artificially inoculated with bacterial blight in the field and susceptible plants are destroyed.

F 2 populations from the individual F 1 plants of top-crosses or double crosses are grown separately and selections are made only from agronomically suitable populations. We grow about 2000 to 5000 F2 plants of each cross combination and 100 to 500 plants are selected from promising combinations for growing in the pedigree nurseries.

Pedigree nurseries. We generally obtain 25 - 30 gram seeds of each plant selection from the F 2 . This sample is divided into 5 - 6 sets of 5 gm each. One set is used for planting the pedigree nursery We grow one single 5 meter long row consisting of 25 plants, for each plant selection. The remaining seed sets are used for testing for resistance to blast, grassy stunt, green leaf110pper, brown planthopper, grain quality, and sometimes for other traits such as drought tolerance or

74

each row. Thus. in gree row till the

We grow about to diseases and msects.

Yield trials. Selected trials. Each season we resistance to the same major are inoculated with sheath screenhouse.

harvested from the bulk harvested .

. One set nursery three selections from each

Data from the ancestral row of new book where the new data are recorded for the reaction of

data from two seasons are available for each for several

year. All of them are screened for resistance

rows are evaluated for in replicated entries. These entries are also screened for

and insects. ln addition, all the entries in the trials in the field and evaluated for stemborer resistance in the

Promising materials with good grain and multiple resistance to diseases and insects are entered in the international nurseries for dissemination of the seed to other programs. The bulk seed of the lines is also to the scientists in the national programs.

Summary In the rice improvement work at IRRJ major efforts are devoted to developing germ

plasm with multiple resistance to major diseases and insects. Screening techniques have been developed, sources of resistance have been identified, genetics of resistance has been investigated, genes for resistance have been transferred to improved plant type background and improved germ plasm with resistance to as many as five major diseases and four insect species has been developed. The breeding methods and procedures employed in our program are discussed.

References 1. ATHWAL D. S., RATHAK, M. D., BACALANGCO, E. H. & PURA, C. D. (1971).

Genetics of resistance to brown planthoppers and green leafhoppers in Oryza sativa L. CropSci. 11, 747-750.

2. CHENG, C. H. & PATHAK M. D. (1972). Resistance to Nephotettix virescens in rice varieties. Jour. Econo. Entom. 65, l 148---1153.

3. International Rice Research Institute. (1977). Annual R.eport for 1976. Los Banos, Philippines.

4. KHUSH, G. S. (1977). Breeding for resistance in rice. Annals New York Acad. Sci.

75

Inheritance of resistance to grassy virus 134-l

Studies on and blast resistance in Nat. Inst. Sci. Ser. D, l'vliscdlaneous Publication No. 1, 58 pp.

7 LAKSHMINARA Y ANA A. & G. S. New genes for resistance the brown 96 ~-1 00.

8. LING, K. C. rice varieties for resistance to tungro disease. In The vims diseases the rice plant. P. 277-291. The Johns Hopkins Press, Baltimore, Maryland.

9. LING, K. C., AGUIERO, V. M. & S. H. (1970). A mass screening method for testing resistance to grassy stunt disease of rice. Plant Dis. Reptr. 56, 565--569.

10. OU, S. H., LING, K. C., KAUFFMAN, H. E. & KHUSH, G. S. (1975). Diseases of upland rice and their control through varietal resistance. In Major Research in Upland Rice. p. 126 -135. International Rice Research Institute, Los Banos, Philippines.

11. OU, S. H., NUQUE, F. L. & SILVA, J.P. 1). Varietal resistance to bacterial blight of rice. Plant Dis. 55, 17-21.

12. PATHAK, M. D. (1972). Resistance to insect pests in rice varieties. In Rice Breeding. p. 325 ~341. International Rice Research Institute. Los Banos, Philippines.

13. PATHAK, M. D., ANDRES, GALACGAC, N. & RAROS, R. (197 Resistance of Rice Varieties to Striped Borers. International Rice Research Institute. Los Banos, Philippines. 69pp.

14. SATYANARAYANAIAH, K. & REDDI, M. V. (1972). Inheritance of resistance to insect gall midge (Pachydiplosis oryzae Wood Mason) in rice. Andhra Agric. J. 19 (l & 2), 1-8.

15. SHASTRY, S. V. S., JOHN, V. T. & SESHU, D. V. (1972). Breeding for resistance to rice tungro virus in India. In Rice Breeding. p. 239-252. International Rice Research Institute, Los Banos, Philippines.

16. SIWI, B. H. & KHUSH, G. S. (1977). New genes for resistance to the green leafhopper inrice. CropSci. 17, 17--20.

Discussion S. Okabe, Japan: It took a long time for the IRRI breeders to establish high-yielding

varieties which have multiple resistance genes. It is highly possible that these varieties might become susceptible clue to changes in disease races or insect biotypes. In order to promptly meet the requirement of new varieties with other type of resistance, do you utilize any special breeding procedure at IRRI?

Answer: Yes, we are now emphasizing the development of varieties and improved germplasm with horizontal resistance. This is particularly the case with resistance to brown planthopper. We are utilizing a process of recurrent selection for accumulating polygenes for resistance from different parents into improved varieties.

H. Fujimaki, Japan: 1. How are you going to cope with the occurrence of new biotypes in various diseases and

insect pests? 2. Do you believe in the existence of stable horizontal resistance to a specific disease or

insect? Answer:· 1. We are incorporating diverse genes for resistance to each disease and insect in our

breeding materials. If materials carrying a specific gene for resistance become susceptible due to the development of races or biotypes, we will utilize other genes which are effective against these.

2. Yes, I do believe in the existence of horizontal resistance. Resistance governed by

76

is stable and effective all the races or may not be however it is moderate. J. T. Ran, India: How do the multi-resistant varieties behave

enCl)Untered any characteristics.

resist;mt varieties have excellent genes for disease and

K. Kawano. Colombia: ft seems that there an the materials at IRRI during the 1960's. What would have tried to incorporate much genetic since the

Answer: It is difficult to answer this The of crosses largely depends upon the breeding getm the

years lRRI did not have as wide germplasm resources as in the later years. Moreover. the objective in the earlier years was the improvement of plant type and potential. However the were greatly modified to include quality, disease and insec't resistance tolerance to environmental stresses. a large number of varieties were used crosses as donors for various traits.